gutterslob wrote:
TitanMech wrote:

Some of the staff at my university are nazi's...

Well, duuhhh

Looks like this thread got Godwinned ...

I basically miss driver support. I have a machine here with GMA500 (Poulsbo) and another with the AMD Radeon 6320 - both are somewhat to intolerably difficult to use with the average Linux distro. Similarly I miss the ability to just connect and go with any bluetooth device. Network Manager has for a couple of years now allowed me to get 3G access thru my phone quite easily but anything beyond that is a struggle. Oh lastly I like how everything keeps working: every time time I 'upgrade' in Linux I find they've fixed one thing but there's a regression somewhere else - like I'm supposed to test every kind of Wifi authentication so I know I'm going to be okay when I'm on the road! And like the author said, sometimes I wish it was possible to just download binaries, unpack them and run the executable* - that's my SOP with Windows these days. (* I do it with Firefox on Linux but that's an exception.)

el_koraco wrote:
Toolz wrote:

Note that testing does not receive security updates from Debian ...

http://secure-testing.debian.net/

Thanks. You learn something new ...

el_koraco wrote:
jerry525 wrote:

I come from Parsix, based on debian testing, but they have their own repo's so each release is a snapshot of testing. Mint debian does something similar I belive. You've probably considered it but Just wonder if this is interesting for #!

I doubt it, papanominal has on numerous occasions mentioned what a great release Squeeze has been (I concur completely), and there seems little point in changing a winning formula (syncing with Debian Testing before the freeze and easing into the new release).

I don't think you got the question. What I think he's asking is that for releases of Crunchbang based on wheezy (while wheezy = testing) would it be possible to have a repo branched off testing - a repo that is 'buffered' or you might say 'moderated' - like Parsix and Mint do. This way the testing repo doesn't periodically get flooding with updates that have not been tested much.

Note that testing does not receive security updates from Debian - I think Parsix and Mint blend this into their 'testing' repos too.

@jerry525: I don't think it's gonna happen as it's a lot of work, and Corenominal isn't that big on delegation.

@anybody: How sensible would it be to change from Debian's testing repo to Mint's?

corenominal wrote:
Toolz wrote:
corenominal wrote:

Any future releases will use the same date identifier. I think it is more useful than using "alpha", "beta" tags, which are kind of meaningless. It was probably a good idea to tag the first couple of Debian based builds as alphas, but now that the build process is sorted, it is not necessary.

Any future pre-releases or any future 'releases'?

I mean is this considered 'release' or 'development'? You know, from the point of view of Distrowatch and write-ups.

It is a release, as in the images have been released and made public. If anyone would like to apply any other labels, that is their choice. smile

Choice, mmm. So all future images released and made public will have that date identifier? This date identifier shouldn't guide people to thinking it's more or less 'release' or 'development', officially?

corenominal wrote:

The "current" Statler release is r20101205,

So it's a 'release'?

corenominal wrote:

Any future releases will use the same date identifier. I think it is more useful than using "alpha", "beta" tags, which are kind of meaningless. It was probably a good idea to tag the first couple of Debian based builds as alphas, but now that the build process is sorted, it is not necessary.

Any future pre-releases or any future 'releases'?

I mean is this considered 'release' or 'development'? You know, from the point of view of Distrowatch and write-ups.

Schadt91 wrote:
Toolz wrote:
Schadt91 wrote:

another forum I am on did something similar last year and never announced the reason why and it caused a lot of drama.

If you're referring to Mint, it was a lot of people asking dumb, redundant questions - 'drama', yes; people being affected, no.

No, I'm not. I don't use Linux Mint and, god willing, I never will.

Okay, loud and clear! smile

jackbang wrote:

Ideally if the server is open-source too, I might rent an AWS instance or similar, so I can do it all myself

A site I use for work is a badly-written ASP.NET thing that uses 40-60k VIEWSTATE variables and doesn't understand gzip. I modified a PHP proxy program so it gzips everything, strips out VIEWSTATE variables, made sense of javascript links, added digest auth, and modified headers of static files to make then cacheable. Saving: 90%. I couldn't survive without it when I'm out in the middle of nowhere connecting through my phone.

Compressing images would be an easy add-on but this solution is a traditional 'CGI' proxy server - not at all 'elegant' - so not good for general surfing. Let me know how you get along as I'm very interested.

Schadt91 wrote:

another forum I am on did something similar last year and never announced the reason why and it caused a lot of drama.

If you're referring to Mint, it was a lot of people asking dumb, redundant questions - 'drama', yes; people being affected, no.

If you're getting your coding hat on then a 'Report' link on the profile page might be a good idea.

You're mostly describing Opera in Turbo mode. I used to use it with great results on my phone. Now the results seem less impressive - I think my ISP has a problem with the server (in Santa Monica, CA).

I haven't checked out your Propel link - I will - but using rsync- type chunking sounds like an enormous challenge. The system would have to store so many deltas for your average website today - blogs, forums, etc.

Caching content is so much more difficult. The powers-that-be don't want it either. Then look at your tpyical CMS and blog packages - written by PHP-philes with next to no knowledge of HTTP protocols - they drive a coach and horses through any caching strategy.

I ran Squid for a long time about six years ago - found I was saving abt 40% bandwidth. I ran again abt 18 months ago and found I was saving much less.

Note that Opera is much better at caching than Firefox. I actually have 6-year-old bug (Bugzilla, Firefox) open here - that concerns caching.

omns wrote:

Anything that reduces the amount of spam I have get rid of is fine by me smile

Actually deleting spam would reduce the amount of spam you would have to get rid of. They spam because they've found that they can, and the amount increases exponentially.

I don't think the 7-day thing would make much impact on the volume of new spam. A seven-day exposure is plenty of reward for the effort required. As I said by PM (no reply) the spammers want their link from:
http://crunchbanglinux.org/forums/
from 'Newest registered user:'
Currently their links stay there for one, two, six or sixteen hours - until a 'real' member signs up - and that's a huge win for them. As soon as the 'real' member signs up ... they sign another up.

Get rid of the  'Newest registered user:' link.

chillicampari wrote:

Edit- toolz, those screenshots on your link look pretty kicky!

I was worried you meant NSFW! smile

I couldn't access the before site and viewed the Google cache.

Just checked now via TOR and yeah, they look alright don't they? smile

So, in the FLOSS spirit, the acceptable workarounds would be: 1) forking (somewhat extreme), or 2) submit a bug report, or 3) patching some kind of theming capability ... perhaps? smile Somebody started on the latter:
http://www.swanson.ukfsn.org/xss/
... five years ago.

Following from the 'toolkits' link there's this:
http://www.jwz.org/xscreensaver/toolkits.html

For a dialog as simple as the password entry box, the "look" of the toolkit really just comes down to colors, fonts, and border widths. Personally, I think the current unlock dialog looks very much like the default GTK theme, but if you disagree, it's easy to tweak it by simply editing the colors and border sizes in the app-defaults file. (If you think you have changes that make it look more "conventional", please send them to me.)

(BTW this thread is a question, not a tip.)

That's just unreasonable.

I mentioned this here:
http://crunchbanglinux.org/forums/topic … questions/

For such an important suggestion this is a somewhat slap-dash opening post. It's almost a troll.

groovydaddy wrote:

... I just wish that the office suite were more complete.  Why doesn't #! come with OO.o instead of Abi and Gnum?

Complete? You really need Database, Presentation and Drawing or are you saying the Writer and Spreadsheet apps need to be more feature-filled?

Note Abiword and Gnumeric are being worked on. Ooo?

For panel: Replace tint2 with lxpanel or xfce panel.

For bluetooth: Don't uninstall - just install BUM (boot up manager) and disable it.

Mich wrote:

Text install will allow computers will less resource to install Debian, which have merit in itself.

Yep. But the suggestion is for 'having'; not 'substituting'.

I guess there'd be no question of taking the text installer away.

Hopefully Debian will do a live installer. Failing that, hopefully the EB4 effort will produce something usable.

19

(10 replies, posted in Help & Support (Stable))

So, looks like the clever ways don't work.

charl wrote:

... if I create a ghost image of my system, could I reinstall windows, do the update, then set the ghost up again like it was?

As anon said, yes. There are several Ghost alternatives - my favourite is partclone - can d/l and run from a live distro.

Next thing to do get on their support channels, forums, whatever, and complain loudly about this.

A good reason for having an installer working from live boot:
http://crunchbanglinux.org/forums/topic … installer/
... wonder how eb4 is coming along?

The second partition will be preserved, as long as you don't format. Deleting the config files and starting afresh is often a good idea, esp if it's a different distro.

22

(11 replies, posted in Tips, Tricks & Scripts)

Might as well! wink

Same on the Eee. Actually I imagine it's a pretty normal circumstance. However you *can* actually proceed.

24

(10 replies, posted in Help & Support (Stable))

That .exe might just be a simple affair that extracts a .bin to a target device - I don't think the .exe is doing anything with the BIOS. I don't know if the usual Linux archiving tools can inspect Win .exe files. An option is to try UniversalExtractor under Wine or a VM.

The last netbook I flashed just wanted a .bin file with a certain name in the root of the the boot partition.

Nah. I think "9.04 will be supported until October" sounds more honest than "9.04 will be fully supported until October".

Re support on forums, 3rd party apps - that will slow down well before or linger past the the mentioned cutoff.